Download Choosing the metabolomics platform

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Proteomics wikipedia , lookup

Sample preparation in mass spectrometry wikipedia , lookup

Nuclear magnetic resonance spectroscopy of proteins wikipedia , lookup

Protein mass spectrometry wikipedia , lookup

Transcript
07/13/2016
Choosing the metabolomics platform
Stephen Barnes, PhD
Department of Pharmacology & Toxicology
University of Alabama at Birmingham
[email protected]
Challenges
• Unlike DNA, RNA and proteins, the metabolome is phenomenally chemically diverse
• Ranges from a gas (H2) that prevades the universe and is the principal component of the Sun
to
• Earwax (long chain fatty acids, both saturated and unsaturated, alcohols, squalene, and cholesterol)
• No single method of analysis
1
07/13/2016
Early forms of metabolomics
2D‐paper chromatogram
low capacity
2D‐Thin layer chromatography of lipids
KO of cerebrosidesulfatase in kidney
These days can be studied by direct electrospray ionization (DESI)
2
07/13/2016
Decision tree
Targeted
Targeted vs
untargeted
NMR or
GC‐MS vs LC‐MS
Extraction method
GC‐MS vs LC‐MS
GC‐MS
LC‐MS
Untargeted
Extraction method
Extraction method
NMR
GC‐MS
LC‐MS
? Capillary‐electrophoresis MS
Metabolomics and GC‐MS
• PROS
– Capillary columns can achieve very high chromatographic resolution
– Retention times are reproducible
– Mass spectral libraries are well developed
• CONS
– Not all compounds can be analyzed by GC‐MS
– Although amino acids, sugars, fatty acids, amines and organic acids can be derivatized, complex polyphenol glycosides and polar lipids are too unstable, even when derivatized, at the temperatures used to elute them
– Approximate mass limit of 400 Da
3
07/13/2016
Two dimensional GC to resolve metabolites
As compounds elute from column 1, they are passed to (cooler) column 2 where they condense. After a period of collection, column 2 is heated so as to separate and elute the compounds.
Leco Corp.
Nuclear Magnetic Resonance (NMR) Spectroscopy • Detects NMR active nuclei • Robust and highly reproducible
• Non‐destructive
• Quantitative
• Used in • Structure elucidation
• Small molecules
• Macromolecules (DNA, RNA, Proteins)
• A number of techniques
• 1D , 2D, 3D
• Molecular motion and dynamics
• Similar method used in medical Imaging (MRI, fMRI)
from Wimal Pathmasiri
4
07/13/2016
NMR considerations
• Sample amount:
• Typical 600 MHz instrument requires 0.5 ml plasma/serum
• Higher field instrument and micro coil detector allows use of 0.1 ml • Quality control:
• In the UK Phenome Center, all samples are analyzed by NMR
• This allows for detection of outliers
• Also found that there is a correlation between the NMR spectrum and whether problems occur in LC‐MS analysis
• NMR analysis used to filter out these samples Liquid chromatography‐Mass Spectrometry
• PROS
• Almost all compounds can be analyzed by LC‐MS
• hydrocarbons do not ionize
• Several orders of magnitude increased sensitivity compared to NMR
• Can collect MS, MSMS and ion mobility data
• CONS
•
•
•
•
Not uniformally quantitative Mass spectral libraries are not well enough developed
Chromatographic separation not adequate
Retention time reproducibility not as good as GC‐MS
5
07/13/2016
The LC
• 1D‐approach
• Use of reverse‐phase, normal phase and HILIC phase
• particle size – smaller is more efficient, but back pressure is a problem Same efficiency, but shorter run times
Higher efficiency
LC flow rate
• Sensitivity is inversely related to flow rate • Slower flow gives more sensitivity
microflow/capillary (5‐50 l/min)
normal flow (0.2‐0.4 ml/min)
nanoflow (0.3‐5 l/min)
6
07/13/2016
Optimizing nanoLC for metabolomics
• Objective is to develop metabolomics for small animal model systems
• D. melangaster
• C. elegans
• D. rerio
• A single zebrafish yields about 1 l of plasma
• Need to move down to the nanoscale
• Important to maintain consistency and quantitation
• Reproducible columns and temperature
Close up of a nanochipLC cartridge (15 cm x 0.2 mm ID). • Each long section of the column is ~2.5 cm (1 inch).
• Can be machined to a better tolerance.
• Simpler connections to the liquid stream.
• Can be placed in a temperature‐
controlled environment
7
07/13/2016
NanoLC‐MS metabolomics
Nanoflex
Reproducibility of peak areas using the nano chipLC approach
10,000,000
Peak area – sample 2
1,000,000
100,000
10,000
1,000
100
100
1,000
10,000
100,000
1,000,000
10,000,000
Peak area – sample 1
8
07/13/2016
Coefficient of variation of the mass of peaks identified by XCMS using nanoLC‐MS
100
mean mass variation = 0.793 ppm for three separate extracts from one sample
variation in mass (ppm)
10
1
0.1
0.01
0.001
0.0001
0
5
10
15
20
25
30
Retention time (min)
Coefficient of variation of retention time for the three separate extracts by nanoLC‐MS Percent variation in retention time
10
mean retention time variation = 0.233%
1
0.1
0.01
0.001
0.0001
0
5
10
15
20
25
30
Retention time (min)
9
07/13/2016
The mass spectrometer
• For untargeted analysis it is important to have high mass resolution and accuracy
• Initial data analysis is performed on the molecular ions
• Each metabolite has a unique mass (m/z) • Nonetheless, a particular mass, however exact, is not necessarily a unique metabolite
• Fourier transform‐ion cyclotron resonance and Orbitrap instruments have the greatest mass accuracy
• However, their performance is time‐dependent and is degraded significantly by short acquisition times (<100 ms)
• They are best used for follow up experiments
Mass analyzer of choice for untargeted metabolomics
• Quadrupole‐orthogonal time‐of‐flight (Q‐TOF)
Agilent 6500
Waters Synapt G2/HMDS
Bruker
Sciex TripleTOF 6600
Current models have ~100,000 mass resolution and 1‐2 ppm mass accuracy
10
07/13/2016
RAMMP, speeding up metabolomics
Gray et al., Anal. Chem., 2016, 88 (11), pp 5742–5751
RAMMP
• There was a reduction in independent features
• 19,000 by conventional method
• 6,000 by RAMMP
Conventional method
RAMMP
Gray et al., Anal. Chem., 2016, 88 (11), pp 5742–5751
11
07/13/2016
Targeted vs untargeted methods
• If we know what the metabolites to be measured are (from previous untargeted analyses, or prior knowledge), then a multiple reaction monitoring (MRM) approach is the best way to go since allows quantitative analysis of possibly 100s of metabolites
• If there is no hypothesis, but instead you want to generate hypotheses, then the untargeted approach is better.
Multiple reaction ion monitoring
LC
Ionizer
Quantitative analysis of metabolites in a complex mixture carried out using a triple quadrupole instrument
Q1 Q2 Q3 Detector
Based on precursor ion/product ion pair(s)
Courtesy, John Cutts
12
07/13/2016
How many MRM transitions?
• Acquisition can be as little as 2 msec, but acquisition time determines sensitivity
• Fast switching electronics can measure as many as 500 different transitions per second
• Since measuring the area under a peak requires 10 data points, the number of transitions measured has to be matched against the shape and width of the chromatographic peaks – to be discussed in more detail later
Combined channels for Krebs cycle
13
07/13/2016
Ion mobility mass spectrometry
• Another method of separating classes of compounds as well as compounds with the same molecular mass
This is a gas‐phase separation of these phospholipids, i.e., no chromatography
A hands‐on session on this will be available as elective on Wednesday afternoon
Waters have a totally different approach to ion mobility – this will be discussed on Thursday by Tom Beaty
Imaging mass spectrometry
Generated by Janusz Kabarowski – a hands‐on elective for 5 people on Wednesday
14
07/13/2016
Questions?
15